McCann, M.E. et al. (2017) Differences in blood pressure...
Transcript of McCann, M.E. et al. (2017) Differences in blood pressure...
McCann, M.E. et al. (2017) Differences in blood pressure in infants after
general anesthesia compared to awake regional anesthesia (GAS Study—a
prospective randomized trial). Anesthesia and Analgesia, 125(3), pp. 837-
845. (doi:10.1213/ANE.0000000000001870)
This is the author’s final accepted version.
There may be differences between this version and the published version.
You are advised to consult the publisher’s version if you wish to cite from
it.
http://eprints.gla.ac.uk/142281/
Deposited on: 31 July 2017
Enlighten – Research publications by members of the University of Glasgow
http://eprints.gla.ac.uk
1
Differences in Blood Pressure in Infants after General Anesthesia compared to Awake
Regional Anesthesia (GAS Study-a prospective randomized trial).
Authors: and The GAS Consortium
1. M.E. McCann1
2. D.E. Withington2, 3
3. S.J. Arnup4
4. A.J. Davidson5,6,7
5. N. Disma8
6. G. Frawley5,6,7
7. N.S. Morton 9,10
8. G. Bell10
9. R.W. Hunt7,11,12
10. D.C. Bellinger13,14,15
11. D.M. Polaner16
12. A. Leo17
13. A.R. Absalom18
14. B.S. von Ungern-Sternberg19,20
15. F. Izzo21
16. P. Szmuk22
2
17. V. Young1
18. S.G. Soriano 1
19. J. C. de Graaff 23, 24,25
20. The GAS Consortium
1. Department of Anaesthesiology, Perioperative and Pain Medicine, Boston Children’s
Hospital, Harvard Medical School, Boston, Massachusetts
2. Department of Anesthesia, Montreal Children’s Hospital, Montreal, Canada
3. Department of Anesthesia, McGill University, Montreal, Canada
4. Clinical Epidemiology and Biostatistics Unit, Murdoch Children’s Research Institute,
Melbourne, Victoria, Australia
5. Anaesthesia and Pain Management Research Group, Murdoch Children’s Research Institute,
Melbourne, Victoria, Australia
6. Department of Anaesthesia and Pain Management, The Royal Children’s Hospital,
Melbourne, Victoria, Australia
7. Department of Paediatrics, University of Melbourne, Melbourne, Victoria, Australia
8. Department of Anaesthesia, Istituto Giannina Gaslini, Genoa, Italy
9. Academic Unit of Anaesthesia , Pain and Critical Care, University of Glasgow, Glasgow,
United Kingdom
10. Department of Anaesthesia , Royal Hospital for Sick Children, Glasgow, United
3
Kingdom
11. Department of Neonatal Medicine, The Royal Children’s Hospital, Melbourne, Victoria,
Australia
12. Neonatal Research Group, Murdoch Children’s Research Institute, Melbourne, Victoria,
Australia
13. Department of Neurology, Boston Children’s Hospital, Harvard Medical School, Boston,
Massachusetts, USA
14. Department of Psychiatry, Boston Children’s Hospital, Harvard Medical School, Boston,
Massachusetts, USA
15. Department of Environmental Health, Harvard T.H. Chan School of Public Health, Boston,
Massachusetts, USA
16. Departments of Anaesthesiology and Pediatrics, Children’s Hospital Colorado, University of
Colorado School of Medicine, Aurora, Colorado, USA
17. Department of Anaesthesia, Royal Children's Hospital, Melbourne, Australia
18. University Medical Center Groningen, Groningen University, The Netherlands
19. Pharmacology, Pharmacy, Anaesthesiology Unit, School of Medicine and Pharmacology,
The University of Western Australia, Perth, Western Australia, Australia
20. Department of Anaesthesia and Pain Management, Princess Margaret Hospital for Children,
Perth, Western Australia, Australia
21. Department of Anaesthesiology and Intensive Care, Paediatric Intensive Care Unit Children
Hospital 'Vittore Buzzi', Milano, Italy
4
22. Department of Anaesthesiology and Pain Management, UT Southwestern and Children’s
Health Medical Center, Dallas, TX and Outcome Research Consortium, Cleveland, OH
23. Department of Anaesthesiology, Wilhelmina Children's Hospital, University Medical Center
Utrecht, Utrecht, the Netherlands
24. Brain Center Rudolph Magnus, University Medical Centre Utrecht, The Netherlands
25. Department of anesthesia, Sophia Children’s Hospital, Erasmus Medical Center Rotterdam,
The Netherlands
Author’s Contribution:
MEM was involved in study design and concept, conduct, data coordination, contribution to
statistical analysis plan, data interpretation, writing and coordinating drafts of the report and
revising it critically, and approving of the version to be published. DW was involved in
study design and conduct, data acquisition and coordination, data interpretation and revising
the report critically. SJA was involved in the development of the statistical analysis plan,
data interpretation and revising the report critically, AJD was involved in study design and
concept, conduct, data coordination, contribution to statistical analysis plan, data
interpretation, writing and coordinating drafts of the report and revising it critically. ND, GF,
NSM, GB, DMP, ARA, BS vUS, FI, PS, VY, SS, JCdG were involved in study design and
conduct, data acquisition and coordination, data interpretation and revising the report
critically. DCB, RWH were involved in study design and conduct, data interpretation and
5
revising the report critically. AL was involved in study conduct, data acquisition and
interpretation and revising the report critically.
26.
Corresponding Author:
Mary Ellen McCann M.D., M.P.H., Department of Anaesthesiology, Perioperative and Pain
Medicine, Boston Children’s Hospital, Harvard Medical School, Boston, Massachusetts
Email Address: [email protected] Phone Number: 617-355-7737
Fax Number: 617-739-0894
Disclosure of Funding:
Declaration of Interests:
The authors declare no competing interests.
Conflicts of Interest: None
Trial Registry Number:
The GAS study is registered in Australia and New Zealand at ANZCTR: ID#
ACTRN12606000441516 first registered on 16th October 2006, Principal Investigators Andrew
6
Davidson, Mary Ellen McCann and Neil Morton; in the USA at ClinicalTrials.gov: ID#:
NCT00756600 first registered on 18th September 2008, Principal Investigators Andrew
Davidson, Mary Ellen McCann and Neil Morton; and in UK at UK Clinical Research Network
(UKCRN) ID#: 6635 (ISRCTN ID#: 12437565; MREC No: 07/S0709/20) Principal Investigator
Neil Morton. The protocol for the GAS study has been previously published by The Lancet.13
Number of words in Abstract: 248
Number of words in Introduction:188
Number of words in Discussion:986
Abbreviated Title: Hypotension after infant spinal or general anesthesia
7
Abstract:
Background: The General Anesthesia compared to Spinal anesthesia (GAS) study is a
prospective randomized, controlled, multi-site, trial designed to assess the influence of general
anesthesia (GA) on neurodevelopment at five years of age. A secondary aim obtained from the
blood pressure data of the GAS trial is to compare rates of intraoperative hypotension after
anesthesia and to identify risk factors for intraoperative hypotension.
Methods: 722 infants ≤ 60 weeks postmenstrual age undergoing inguinal herniorrhaphy were
randomized to either bupivacaine RA or sevoflurane GA. Exclusion criteria included risk factors
for adverse neurodevelopmental outcome and infants born < 26 weeks’ gestation. Moderate
hypotension was defined as mean arterial pressure (MAP) measurement of < 35 mm Hg. Any
hypotension was defined as mean arterial pressure (MAP) of <45 mm Hg. Epochs were defined
as a 5 minute measurement periods. The primary outcome was any measured hypotension <35
mm Hg from start of anesthesia to leaving the operating room. This analysis is reported
primarily as intention to treat (ITT) and secondarily as per protocol (APP).
Results: The relative risk of general anesthesia compared with regional anesthesia predicting any
measured hypotension<35 mm Hg from the start of anesthesia to leaving the operating room was
2.8 (CI 2.0, 4.1, p value <0.001) by ITT analysis and 4.5 (CI 2.7,7.4, p value <0.001) by APP
analysis(?).
8
In the GA group 87% and 49% and in the RA group 41% and 16% exhibited any or moderate
hypotension by ITT, respectively. In multivariable modeling, group assignment (GA vs. RA),
weight at the time of surgery and minimal intraoperative temperature were risk factors for
hypotension. Interventions for hypotension occurred more commonly in the GA group
compared with the RA group (Relative Risk 2.8, 95% CI:1.7,4.4 by ITT).
Conclusions: RA reduces the incidence of hypotension and chance of intervention to treat it
compared with sevoflurane anesthesia in young infants undergoing inguinal hernia repair.
Keywords: Anesthesia, General; Anesthesia, Spinal; Blood Pressure; Infant
Introduction:
It is estimated that over 6 million general anesthetics are administered to young children
worldwide annually.(1,2) Several studies have demonstrated an association between surgery in
infancy and increased risk of poor neurobehavioral outcome, although the reason for this
9
association is unclear.(3-5) Since many general anesthetics cause neurotoxicity in young
animals, there is concern that general anesthetics maybe neurotoxic to infants.(6-8) Apart from
neurotoxicity, there may be other modifiable peri-operative factors which impact on long term
neurocognitive outcome, among them hypotension.(9)
The General Anesthesia compared to Spinal anesthesia (GAS) study, a prospective randomized
equivalence trial, which was designed to determine whether GA and RA have similar long-term
effects on the developing brain.{Davidson, 2016 #3031}In this study 722 infants undergoing
inguinal herniorrhaphy were randomized to regional anesthesia (RA) or general anesthesia (GA).
The apnea, interim 2 year neurocognitive and summary blood pressure data have been recently
published elsewhere. The hypothesis of this study is that moderate hypotension is more common
in infants undergoing general anesthesia compared to infants undergoing regional anesthesia as
measured by the incidences of moderate hypotension. Secondary aims were to compare duration
of hypotension and the incidence of interventions to treat it, and identify factors associated with
hypotension.
Methods:
10
This manuscript adheres to the applicable Equator guidelines. The GAS study is registered in
Australia and New Zealand at ANZCTR: ID# ACTRN12606000441516 first registered on 16th
October 2006, Principal Investigators Andrew Davidson, Mary Ellen McCann and Neil Morton;
in the USA at ClinicalTrials.gov: ID#: NCT00756600 first registered on 18th September 2008,
Principal Investigators Andrew Davidson, Mary Ellen McCann and Neil Morton; and in UK at
UK Clinical Research Network (UKCRN) ID#: 6635 (ISRCTN ID#: 12437565; MREC No:
07/S0709/20) Principal Investigator Neil Morton. The protocol for the GAS study has been
previously published by The Lancet.13
Study participants
After institutional review board approval from each site and written informed consent from
parents or guardians, infants were enrolled. (Table 1) Recruitment began in 2007 and ended in
2013. Eligibility criteria included infants up to 60 weeks’ postmenstrual age (PMA) undergoing
inguinal herniorrhaphy (with or without circumcision) born at greater than 26 weeks’ gestation.
Exclusion criteria included contraindications for either anesthetic technique, history of heart
disease requiring surgery or pharmacotherapy, mechanical ventilation immediately prior to
surgery, known chromosomal or congenital abnormalities known to affect neurodevelopment,
previous exposure to volatile GA or benzodiazepines as a neonate or in the third trimester in
11
utero, known neurologic injury such as cystic peri-ventricular leukomalacia or grade three or
four intra-ventricular hemorrhage, and social or geographic factor or language barriers that made
follow up difficult. Eligible infants were identified from operating room schedules and pre-
admission clinics.
Randomization and blinding
A 24-hour web-based randomization was managed by The Data Management & Analysis Centre,
Department of Public Health, University of Adelaide, South Australia. Children were
randomized with a 1:1 allocation ratio to either RA or GA by random permuted blocks of two or
four and stratified by site and gestational age at birth: 26 to 29 weeks and 6 days, 30 to 36 weeks
and 6 days, and 37 weeks and more.
Procedures
Patients in the RA arm received regional anesthesia: either spinal, spinal with caudal, spinal with
ilioinguinal nerve block, or caudal block. The anesthetic used was bupivacaine or
levobupivacaine. Some patients received caudal chloroprocaine intra-operatively to prolong the
block. In the RA arm, any sedation or GA given was considered a protocol violation. Oral
sucrose drops were permitted in the RA arm and acetaminophen (paracetamol) in both arms.
Those in the GA arm received sevoflurane in air/oxygen for induction and maintenance along
with neural blockade via caudal or ilioinguinal block with bupivacaine or levobupivacaine.
Airway support and use of neuromuscular blocking agents was not standardized. No opioids or
12
nitrous oxide were allowed intra-operatively. Blood pressure, heart rate, oxygen saturation, end-
tidal CO2 and temperature were recorded every 5 minutes intra-operatively. Because there was
no airway management standardization within this study, end tidal CO2 measurements were
considered unreliable. These measurements were recorded on the Case Report Forms by a
research assistant and in the anesthesia record. A blood glucose measurement was obtained
perioperatively.
The pre-specified primary outcome for this analysis was at least one epoch of observed moderate
hypotension. Secondary outcomes include observation of at least one epoch of observed any
hypotension; at least three epochs of moderate hypotension, at least three epochs of any
hypotension and average intraoperative MAP as well as interventions done for blood pressures
less than 80% of baseline. Moderate hypotension and any hypotension was defined as single or
multiple measures of MAP of <35 mm Hg and MAP of <45 mm Hg, respectively. NIRS and
Doppler evidence demonstrates that cerebral blood flow decreases at MAPs <45 mm Hg and
cerebral oxygenation decreases at MAPs <35 mm Hg in infants <6 months undergoing
sevoflurane anesthesia.(10) An epoch was defined as a single intraoperative measurement
representing a 5 minute period. Multiple measures within a 5 minute period were averaged.
Mean arterial pressures were calculated from the formula ((systolic BP) +2(diastolic BP))/3. Cuff
position (calf vs arm) was not adjusted for because there is less than a 1 mm Hg difference
between calf and arm pressures in infants less than 3 months.(11,12) Cuff size was determined
by choice of a cuff width that was between 2/3 the length of either the humerus, femur or tibia
13
depending on the limb. Baseline blood pressure measurements were the first blood pressure
measured in the operating room. Intraoperative and PACU interventions for blood pressure
greater than 20% below baseline included lactated Ringer’s bolus of 20 milliliters/kilogram and
vasoactive medications at the discretion of the anesthesiologist.
Prolonged hypotension was defined as 3 or more consecutive epochs of MAP <35 mm Hg.
Any hypotension and moderate hypotension rates were analyzed in three time periods: anesthesia
time: time from the start of the anesthesia to time of leaving the operating room (primary
analysis); pre-incision time: time from the start of anesthesia to incision and
surgical time: time between knife to skin and last stitch. The level of intervention for
intraoperative hypotension was also noted. A significant intervention was defined a-priori as a
fluid bolus or the administration of a vasoactive substance.
Statistical Analysis:
Analysis populations
The primary analysis is reported as intention to treat (ITT) excluding participants who withdrew
consent or were randomized after surgery. A secondary analysis was performed as per-protocol
(APP), which excludes cases where surgery was cancelled, and in the RA arm, any child who
received any sevoflurane or sedative medication.
14
Data analysis
For occurrence of moderate and any hypotension, a comparison between GA and RA groups is
presented as a relative risk (RR) estimated using the generalised estimating equation (GEE)
approach, assuming binomially distributed outcome data, and applying a log link function and
robust standard errors in Stata 13 (Stata Corp LP., USA). A fixed effect was fitted for each
group and gestational age (treated as a continuous outcome) and an exchangeable correlation was
assumed between observations within the same site. It is noted in the results where the GEE
approach did not converge, and an alternative, more conservative method was applied.
For average MAP, a comparison between GA and RA groups is presented as a difference in
means as estimated from a 3-level random intercept model, where site is the top level, individual
is the second level and observation within individual is the third level, using xtmixed with
maximum likelihood estimation in Stata 13 (Stata Corp LP., USA). A random intercept is fit to
each randomization site and individual, and a fixed effect is fit for each group and gestational
age. The residual errors within each individual were assumed to have an autoregressive structure
of order 4, to account for the expected autocorrelation between successive blood pressure
measurements. An order of 4 was chosen by performing a likelihood ratio test between models
fit with successive orders of autogression for observations in the anesthesia time period, and
selecting the highest order where p<0.05. All estimates are presented with 95% confidence
intervals (CIs) and two-sided p-values As a further secondary analysis we aimed to identify
15
risk factors for prolonged hypotension (3 or more consecutive epochs <35 mm Hg). Prolonged
hypotension was chosen instead of any hypotension because the outcomes are likely to be more
severe. . The following risk factors for prolonged hypotension (3 or more consecutive epochs
<35 mm Hg) were identified a-priori: chronological age, gender, weight at time of surgery,
postmenstrual age at time of surgery, gestational age at birth, preoperative fasting time,
preincisional times, intraoperative temperature (mean and minimum values) and duration of
surgery. These risk factors were chosen by previous research and consensus by and clinical
expertise of the authors.
To assess the strength of association of each risk factor with prolonged hypotension we present
estimates from: a tri-variable model as described in Data Analysis, including allocated study
group and gestational age in addition to the risk factor of interest; and a full multivariable model
containing all risk factors of interest. Prior to constructing the full multivariable model we
assessed the set of risk factors for multicollinearity by visual inspection of twoway scatterplots
and by calculating the variance inflation factors. We determined that only one factor in the
following sets of factors could be estimated in the full model: mean intraoperative temperature
and minimum intraoperative temperature; and weight at surgery, PMA at surgery and
chronological age at surgery. We selected the factor with the strongest tri-variable association for
the full model.
16
Sample size considerations
The sample size for the GAS study was based on the 5 year neurodevelopmental outcome.(13)
In line with CONSORT recommendations, post-hoc power calculations were not done and
instead results are presented with confidence intervals, which capture the uncertainty in our
findings that reflect the sample size.
For our sample size of 709, the observed reduction in moderate hypotension from 49% to 16% in
the GA and RA arms respectively can be estimated with a 95% confidence interval of 2.4 to 4.1
for relative risk, using an unadjusted two-sample χ2 proportions test. The confidence intervals
presented in the Results differ because our analysis accounts for the randomization stratification
factors, and non-independence between observations taken within the same site.
Results:
722 infants were recruited into the trial (see figure 1). For the ITT analysis 355 were in the RA
arm and 356 in the GA arm (figure 1). Baseline, demographic, anesthetic and surgical data are
summarized in table 1. There were 394 premature infants and 325 term infants. In the RA arm 70
patients had a protocol violation involving exposure to sevoflurane or sedation; 10 patients were
17
converted to GA with no block attempted, 23 had a partially successful RA but required sedation
after incision and 37 had a regional block attempted but were converted to GA before incision.
Thus for the APP analysis 286 were in the RA arm and 356 in the GA arm. There was no
hypoglycemia noted in any of the infants. There were no infants with patent ductus arteriosus or
other conditions known to alter blood pressure at the time of surgery. Most of the GA group also
received concurrent caudal blocks for intra and postoperative analgesia. The mean end-
expiratory sevoflurane concentration measured in the GA group was 2.6 (0.7) %.
MAP was not measured in 14% of epochs in the RA group and 15% in the GA group. The
missing data occurred mostly at the beginning and end of cases. MAP was missing for 32% of
the first three epochs after the start of anesthesia in both the GA and RA groups, while in the last
epoch 33% of GA and 23% of RA cases were missing MAP. Furthermore, in 71% of GA cases
the first MAP measurement was taken in the 10 minutes after the start of anesthesia, compared to
51% of RA cases. The initial baseline MAP for the GA group was 59 mm Hg and 62 mm Hg for
the RA group by ITT and 61 mm Hg for the RA group by APP.
The relative risk of general anesthesia compared with regional anesthesia predicting any
measured hypotension<35 mm Hg from the start of anesthesia to leaving the operating room was
2.8 (CI 2.0, 4.1, p value <0.001) by ITT analysis and 4.5 (CI 2.7,7.4, p value <0.001) by APP
(Table 4). The percentages of infants in the GA group who exhibited any epochs of any and
moderate hypotension were 87% and 49% respectively. In the RA group the percentages of
18
infants exhibiting any epochs of any and moderate epochs of hypotension were 41% and 16 % by
ITT and 34% and 9% by APP, respectively (Table 2). In the GA group, 15% (54 infants)
experienced 5-9 epochs and 8% (28 infants) experienced 10 or more epochs of moderate
hypotension with three infants experiencing 15 or more epochs. In the RA group on APP
analysis, 8% (22 infants) experienced 1-4 epochs and 2% (5 infants) experienced 5-9 epochs of
moderate hypotension. These epochs could be consecutive or nonconsecutive. In the GA group,
18% (63 infants) had prolonged hypotension (3 or more consecutive measurements<35 mm Hg)
and in the RA group, 3% (11 infants) by ITT and 0% (1 infant) by APP had prolonged
hypotension.
There was strong evidence that the mean MAP for the GA group was 10 mm Hg below the mean
MAP for the RA group in each time period (preincision, surgical time, anesthesia time) and type
of analysis (ITT, APP)(Table 3). There was strong evidence for increased risk of any
hypotension during the anesthesia time period in the GA group in both the ITT population (RR =
2.1, 95% CI:1.8 – 2.4 , p <0.001 ), and the APP population (RR = 2.5, 95% CI:2.1 - 3.0, p
<0.001 ), and similarly for moderate hypotension (ITT, RR = 4.8, 95% CI:2.0 - 4.1, p <0.001 ;
APP, RR = 4.5, 95% CI:2.7 - 7.4, p <0.001) (table 4).
In a multivariable prediction model (N=609, 69 events with complete data) the risk of prolonged
hypotension was significantly associated with assignment to the GA group (GA RR = 4.63, 95%
CI:2.23 - 9.66, p <0.001), lower weight at time of surgery (per kg RR =0.56 , 95% CI :0.47 -
19
0.67, p <0.001) and minimum intraoperative temperature (per deg C RR =0.78 , 95% CI:0.65 -
0.95, p =0.011). There was no evidence for an association between fasting time, surgical time,
and gender (table 5). We note that mean intraoperative temperature and minimum intraoperative
temperature; and weight at time of surgery, PMA and chronological age at surgery, were highly
associated with each other. As a result, the full model can only estimate the association between
one of these factors with prolonged hypotension. The alternative factors provided almost the
same strength of association with prolonged hypotension. Each factor may be independently
important, but the sample size was too small to estimate these associations reliably.
Interventions for hypotension occurred more commonly in the GA group compared with the RA
group (19.2% vs 7.3% by ITT and 19.2% vs 6.6% by APP). The most common intervention
was a fluid bolus, which accounted for 75% or more of the interventions (table 6). The relative
risk for GA compared to RA predicting an intervention for hypotension were 2.8 (95% CI:1.7 -
4.4) by ITT and 2.8 (95% CI:1.7 - 4.9) by APP.
20
Discussion:
In this analysis we found more frequent hypotension in the GA group compared with the RA
group. The GA group was more likely to demonstrate any and moderate hypotension for both
single measurements and 3 or more consecutive measurements.
There are many different working definitions of hypotension for awake premature and term
infants but the acceptable lower limits of blood pressures in neonates and young infants
undergoing general anesthesia have not been established.(14,15) Normative data from preterm
and term infants reveal a steady increase in blood pressure over the first month of life.(12,16-18)
In term infants, the systolic blood pressure at birth averages 62 mm Hg but by one month of age
it is >80 mm Hg, representing an increase of almost 30% .(17) The slope for the rise in blood
pressure for preterm infants is similar.(18) The rise in calculated MAP over the first month of
life is approximately 15% from a mean calculated MAP of 52 mm Hg on day 1 to 61 mm Hg on
day 30.(16,17) It is also important to note that the blood pressures of infants born prematurely
are higher than those of term infants at 41 weeks post menstrual age.{Tan, 1988 #182} The
21
reasons for this are unclear but may have to do with the inflammatory stress that premature
infants endure while developing. These expected blood pressure differences between term and
expremature infants highlight the importance of obtaining accurate baseline blood pressures in
all infants and neonates before anesthesia.
It is generally believed that maintaining blood pressure within the limits of cerebral
autoregulation is optimal for cerebral protection. Several studies have shown that the lower limit
of cerebral autoregulation for some infants is indeed fairly close to the definition of MAP
hypotension using the infant’s age in postmenstrual weeks, although there is also evidence that
some premature infants are able to demonstrate cerebral autoregulation at a MAP level
considerably lower than their gestational age in weeks.(19,20) A retrospective analysis of infants
less than 6 months of age undergoing sevoflurane anesthesia found that the lower limit of
autoregulation was 45 mm Hg but cerebral oxygenation did not decrease until the MAP was less
than 35 mm Hg.(10)
Although it is known that hypotension can lead to decreased cerebral perfusion and thus to
hypoxic ischemic injury to the brain, there are many unknowns to consider in infants undergoing
general anesthesia. First the duration of hypotension that puts infants at risk is unknown. In a
small case series of infants less than 3 months of age undergoing general anesthesia between
120 and 180 minutes who did show evidence of perioperative hypoxic ischemic injury, 5 of the 6
22
patients had a mean intraoperative MAP of less than 35 mm Hg for the entire procedure.(21) It
is important to note as previously published, no neurocognitive differences were found in the 2
year interim neurocognitive evaluations between the GA and RA group even though the lowest
mean single systolic blood pressures noted in the GA group were below the American Heart
Association guidelines for hypotension in neonates at 55 mm Hg.(22,23) Also, studies have
shown that a MAP below 35 mm Hg can lead to decreased cerebral perfusion and cerebral
oximetry values but there has been no established link between these factors intraoperatively and
later neurocognitive outcomes.{Rhondali, 2013 #2695}{Rhondali, 2015 #2693} Risk factors
for poor neurocognitive outcomes including hypotension and duration of hypotension will be
analyzed when the final 5 year neurocognitive outcome assessments are completed.
Other risk factors that were associated with moderate prolonged hypotension were weight at the
time of surgery and minimal temperature recorded during the procedure but these effects sizes
were small. Weight was a stronger predictor of moderate hypotension than chronological age,
postmenstrual age or gestational age at birth. Speculatively, weight might be a proxy for overall
health as well as age.
Criteria for either the timing or type of intervention for hypotension in young infants have not yet
been developed. In our study, a fluid bolus was administered in approximately 75% of
hypotensive episodes; vasoactive agents were the sole therapy in approximately 15% of
interventions in the GA group and 21% in the RA group.
23
Study Limitations:
There are several limitations to this study that warrant discussion. The blood pressure values
were missing in 14% of our study data, most commonly at the very beginning or end of the case.
However, sensitivity analysis of these missing values to include additional hypotensive
measurements for the missing values did not alter the direction of the values. The initial
measured blood pressures in the RA group occurred later than the initial blood pressure
measurements in the GA group, which might have introduced a bias.
Another limitation is that location (calf vs arm) and sizes were not standardized as part of the
protocol. Pediatric anesthesiologists conventionally use the rule of 2/3rds; the cuff width should
be 2/3 the length of the underlying bone of the limb measured. However, this is not the most
accurate method of sizing blood pressure cuffs. Measuring the limb circumference at midpoint
and choosing a blood pressure cuff width that is 40% of this circumference is more
accurate.(24,25) This may have introduced some nonsystematic bias into the data since the
participants are likely to follow the same rule in all patients. The cuff position among patients in
the RA group was predominantly the lower limbs. We chose not to adjust for limb placement
because there is very little difference in limb blood pressure measurements in infants. However,
leg blood pressures are usually slightly lower than arm blood pressures so any bias in our
measurements would result in underestimates of the true blood pressure in the regional group.
24
There also was no standardization of the noninvasive blood pressure monitoring systems across
the sites however both these factors should have been accounted for by randomization.
Another limitation is that the definitions of hypotension in infants undergoing general anesthesia
are not really developed. The blood pressures we considered to indicate hypotension may in fact
have been adequate blood pressures for some or all of the infants who exhibited them. In
addition, the risks of hypotension to neurodevelopment are presumably duration dependent and
“at risk” duration is not known. We arbitrarily chose 3 or more consecutive epochs of moderate
hypotension to be important.
In conclusion, a substantial percentage of our patients had calculated mean arterial blood
pressures less than 35 mm Hg at some point during their anesthetic and the incidence was much
higher for those who received sevoflurane anesthesia. The importance and consequences,
particularly on neurodevelopment, of transient intraoperative hypotension during anesthesia in
infancy are unknown. The potential for interplay between hypotension and the putative direct
effects of general anesthetics on neurodevelopment are especially intriguing. Further
consideration and investigation of this issue is warranted.
Acknowledgements:
We would like to acknowledge all of these collaborators for their assistance and advice with the
GAS study: Mark Fajgman MBBS, FANZCA Department of Anaesthesia, Monash Medical
25
Centre, Melbourne, Australia; Daniela Tronconi RN Department of Anaesthesia, Istituto
Giannina Gaslini, Genoa, Italy; David C van der Zee MD, PhD Department of Pediatric Surgery,
Wilhelmina Children’s Hospital, University Medical Centre Utrecht, The Netherlands; Jan BF
Hulscher MD, PhD Department of Surgery, University Medical Center Groningen, Groningen
University, Groningen, The Netherlands; Rob Spanjersberg RN Department of Anaesthesiology,
University Medical Center Groningen, Groningen University, Groningen, The Netherlands;
Michael J Rivkin MD Department of Neurology, Boston Children's Hospital, Boston,
Massachusetts; Michelle Sadler-Greever RN Department of a and Pain Medicine, University of
Washington, Seattle, Washington and Department of Anaesthesia and Pain Medicine, Seattle
Children's Hospital, Seattle, Washington; Debra Faulk MD Department of Anaesthesiology,
Children's Hospital Colorado, Denver, Colorado and Department of Anaesthesiology, University
of Colorado School of Medicine, Denver, Colorado; Danai Udomtecha MD, Sarah Titler MD,
Susan Stringham RN, Pamela Jacobs RN and Alicia Manning AND Department of Anaesthesia,
The University of Iowa Hospital and Clinics, Iowa City, Iowa; Roxana Ploski BS and Alan
Farrow-Gillespie MD Department of Anaesthesiology, Children’s Medical Center Dallas, Dallas,
Texas and Department of Anaesthesiology, University of Texas Southwestern Medical Center,
Dallas, Texas and Dallas and Children’s Medical Center at Dallas and Outcome Research
Consortium, Dallas, Texas; Timothy Cooper MA, PsyD Division of Developmental Medicine &
the Center for Child Development, Monroe Carell Jr Children’s Hospital at Vanderbilt,
Nashville, Tennessee; Elizabeth Card MSN, APRN, RN, FNP-BC, CPAN, CCRP Perioperative
Clinical Research Institute, Vanderbilt University Medical Center, Nashville, Tennessee; Wendy
26
A Boardman BA Department of Anaesthesiology, Dartmouth-Hitchcock Medical Center,
Lebanon, New Hampshire; Theodora K Goebel RN, BSN, CCRC Department of
Anaesthesiology and Critical Care Children's Hospital of Philadelphia, Philadelphia,
Pennsylvania.
Declaration of Interests:
The authors declare no competing interests.
Funding Institutions:
All hospitals and centers were generously supported by anaesthesiology departmental
funding. In addition to this funding, specific grants received for this study are as follows:
Australia and New Zealand: The Australian National Health & Medical Research Council,
Canberra, Australian Capital Territory, Australia; Australian and New Zealand College of
Anaesthetists, Melbourne, Victoria, Australia; Murdoch Children’s Research Institute,
Melbourne, Victoria, Australia. This study was supported by the Victorian Government’s
Operational Infrastructure Support Program in Melbourne, Victoria, Australia; Department of
Anaesthesia and Pain Management, Royal Children’s Hospital, Melbourne, Victoria,
Australia; Department of Anaesthesia , Monash Medical Centre, Melbourne, Victoria,
Australia; Department of Anaesthesia and Pain Management, Princess Margaret Hospital
Foundation, Perth, Western Australia, Australia; Department of Paediatric Anaesthesia ,
Women's Children's Hospital, Adelaide, South Australia, Australia and Department of
Paediatric Anaesthesia and Operating Rooms, Starship Children's Hospital, Auckland, New
27
Zealand. United States: National Institute of Health, Bethesda, Maryland, United States;
Food and Drug Administration, Silver Spring, Maryland, United States; Department of
Anaesthesiology, Perioperative and Pain Medicine, Boston Children’s Hospital, Boston,
Massachusetts; Department of Anaesthesia and Pain Medicine, Seattle Children's Hospital,
Seattle, Washington, United States; Department of Anaesthesiology, Children’s Hospital
Colorado, Denver, Colorado, United States; Department of Anaesthesia, University of Iowa
Hospital , Iowa City, Iowa, United States; Department of Anaesthesiology, Children’s
Medical Center Dallas, Dallas, Texas, United States; Department of Pediatric
Anaesthesiology, Anne and Robert H. Lurie Children’s Memorial Hospital, Chicago, Illinois,
United States; Department of Anaesthesiology, Dartmouth Hitchcock Medical Center.
28
References
1. Weiser TG, Regenbogen SE, Thompson KD, Haynes AB, Lipsitz SR, Berry WR,
Gawande AA. An estimation of the global volume of surgery: a modelling strategy based
on available data. Lancet 2008;372:139-44.
2. Tzong KY, Han S, Roh A, Ing C. Epidemiology of pediatric surgical admissions in US
children: data from the HCUP kids inpatient database. J Neurosurg Anesthesiol
2012;24:391-5.
3. Wilder RT, Flick RP, Sprung J, Katusic SK, Barbaresi WJ, Mickelson C, Gleich SJ,
Schroeder DR, Weaver AL, Warner DO. Early exposure to anesthesia and learning
disabilities in a population-based birth cohort. Anesthesiology 2009;110:796-804.
4. DiMaggio C, Sun LS, Li G. Early childhood exposure to anesthesia and risk of
developmental and behavioral disorders in a sibling birth cohort. Anesth Analg
2011;113:1143-51.
5. Ing C, Dimaggio C, Whitehouse A, Hegarty MK, Brady J, von Ungern-Sternberg BS,
Davidson A, Wood AJ, Li G, Sun LS. Long-term Differences in Language and Cognitive
Function After Childhood Exposure to Anesthesia. Pediatrics 2012.
6. Jevtovic-Todorovic V, Hartman RE, Izumi Y, Benshoff ND, Dikranian K, Zorumski CF,
Olney JW, Wozniak DF. Early exposure to common anesthetic agents causes widespread
neurodegeneration in the developing rat brain and persistent learning deficits. J Neurosci
2003;23:876-82.
7. Stratmann G, Sall JW, May LD, Bell JS, Magnusson KR, Rau V, Visrodia KH, Alvi RS,
Ku B, Lee MT, Dai R. Isoflurane differentially affects neurogenesis and long-term
neurocognitive function in 60-day-old and 7-day-old rats. Anesthesiology 2009;110:834-
48.
8. Yon JH, Daniel-Johnson J, Carter LB, Jevtovic-Todorovic V. Anesthesia induces
neuronal cell death in the developing rat brain via the intrinsic and extrinsic apoptotic
pathways. Neuroscience 2005;135:815-27.
9. Englehart MS, Allison CE, Tieu BH, Kiraly LN, Underwood SA, Muller PJ, Differding
JA, Sawai RS, Karahan A, Schreiber MA. Ketamine-based total intravenous anesthesia
versus isoflurane anesthesia in a swine model of hemorrhagic shock. The Journal of
trauma 2008;65:901-8; discussion 8-9.
29
10. Rhondali O, Pouyau A, Mahr A, Juhel S, De Queiroz M, Rhzioual-Berrada K, Mathews
S, Chassard D. Sevoflurane anesthesia and brain perfusion. Paediatr Anaesth
2015;25:180-5.
11. Crapanzano MS, Strong WB, Newman IR, Hixon RL, Casal D, Linder CW. Calf blood
pressure: clinical implications and correlations with arm blood pressure in infants and
young children. Pediatrics 1996;97:220-4.
12. Park MK, Lee DH. Normative arm and calf blood pressure values in the newborn.
Pediatrics 1989;83:240-3.
13. Davidson AJ, Morton NS, Arnup SJ, de Graaff JC, Disma N, Withington DE, Frawley G,
Hunt RW, Hardy P, Khotcholava M, von Ungern Sternberg BS, Wilton N, Tuo P, Salvo
I, Ormond G, Stargatt R, Locatelli BG, McCann ME, General Anesthesia compared to
Spinal anesthesia C. Apnea after Awake Regional and General Anesthesia in Infants: The
General Anesthesia Compared to Spinal Anesthesia Study--Comparing Apnea and
Neurodevelopmental Outcomes, a Randomized Controlled Trial. Anesthesiology
2015;123:38-54.
14. Brady KM, Mytar JO, Lee JK, Cameron DE, Vricella LA, Thompson WR, Hogue CW,
Easley RB. Monitoring cerebral blood flow pressure autoregulation in pediatric patients
during cardiac surgery. Stroke 2010;41:1957-62.
15. Nafiu OO, Voepel-Lewis T, Morris M, Chimbira WT, Malviya S, Reynolds PI, Tremper
KK. How do pediatric anesthesiologists define intraoperative hypotension? Paediatr
Anaesth 2009;19:1048-53.
16. Tan KL. Blood pressure in full-term healthy neonates. Clin Pediatr (Phila) 1987;26:21-4.
17. Report of the Second Task Force on Blood Pressure Control in Children--1987. Task
Force on Blood Pressure Control in Children. National Heart, Lung, and Blood Institute,
Bethesda, Maryland. Pediatrics 1987;79:1-25.
18. Tan KL. Blood pressure in very low birth weight infants in the first 70 days of life. J
Pediatr 1988;112:266-70.
19. Tyszczuk L, Meek J, Elwell C, Wyatt JS. Cerebral blood flow is independent of mean
arterial blood pressure in preterm infants undergoing intensive care. Pediatrics
1998;102:337-41.
20. Munro MJ, Walker AM, Barfield CP. Hypotensive extremely low birth weight infants
have reduced cerebral blood flow. Pediatrics 2004;114:1591-6.
21. McCann ME, Schouten AN, Dobija N, Munoz C, Stephenson L, Poussaint TY, Kalkman
CJ, Hickey PR, de Vries LS, Tasker RC. Infantile postoperative encephalopathy:
perioperative factors as a cause for concern. Pediatrics 2014;133:e751-7.
22. Davidson AJ, Disma N, de Graaff JC, Withington DE, Dorris L, Bell G, Stargatt R,
Bellinger DC, Schuster T, Arnup SJ, Hardy P, Hunt RW, Takagi MJ, Giribaldi G,
Hartmann PL, Salvo I, Morton NS, von Ungern Sternberg BS, Locatelli BG, Wilton N,
Lynn A, Thomas JJ, Polaner D, Bagshaw O, Szmuk P, Absalom AR, Frawley G, Berde
C, Ormond GD, Marmor J, McCann ME, consortium GAS. Neurodevelopmental
30
outcome at 2 years of age after general anaesthesia and awake-regional anaesthesia in
infancy (GAS): an international multicentre, randomised controlled trial. Lancet 2015.
23. Kleinman ME, Chameides L, Schexnayder SM, Samson RA, Hazinski MF, Atkins DL,
Berg MD, de Caen AR, Fink EL, Freid EB, Hickey RW, Marino BS, Nadkarni VM,
Proctor LT, Qureshi FA, Sartorelli K, Topjian A, van der Jagt EW, Zaritsky AL,
American Heart A. Pediatric advanced life support: 2010 American Heart Association
Guidelines for Cardiopulmonary Resuscitation and Emergency Cardiovascular Care.
Pediatrics 2010;126:e1361-99.
24. Devinck A, Keukelier H, De Savoye I, Desmet L, Smets K. Neonatal blood pressure
monitoring: visual assessment is an unreliable method for selecting cuff sizes. Acta
Paediatr 2013;102:961-4.
25. Clark JA, Lieh-Lai MW, Sarnaik A, Mattoo TK. Discrepancies between direct and
indirect blood pressure measurements using various recommendations for arm cuff
selection. Pediatrics 2002;110:920-3.
31
Table 1. Baseline demographic anesthetic and surgical data.
32
Demographics RA arm as intention
to treat N=361
GA arm as intention
to treat N=358
RA arm as per
protocol N=286
Male gender 294 (81.7%) 306 (85.5%) 231 (80.8%) Gestational age at birth
(days) 248.3 (28.5) 248.6 (27.2) 248.2 (28.7)
Chronological age at surgery
(days)* 70.1 (31.8) 71.0 (31.7) 68.9 (30.9)
Post menstrual age at surgery
(days)* 318.3 (32.6) 319.5 (32) 317.1 (32.0)
Birth weight (kg) 2.4 (0.9) 2.3 (0.9) 2.3 (0.9) Weight at time of surgery
(kg)* 4.2 (1.1) 4.3 (1.1) 4.2 91.1)
Median Apgar at 1 minute 9 (7-9) 9 (7-9) 9 (7-9) Median Apgar at 5 minutes 9 (9-10) 9 (9-10) 9 (9-10) Ever ventilated with a
tracheal tube 47 (13.1%) 45 (12.6%) 37 (12.9%)
Ever required supplemental
oxygen 95 (26.4%) 81 (22.6%) 76 (26.6%)
Supplemental oxygen
immediately prior to
surgery*
6 (1.7%) 6 (1.7%) 4 (1.4%)
Electronic monitoring for
apnea in previous 24 hr 17 (4.7%) 17 (4.8%) 13 (4.6%)
Observed apnea previous 24
hrs * 6 (1.7%) 8 (2.2%) 6 (2.1%)
Fasting time (mins)* 368.2 (146.4) 367.3 (155.1) 370.7 (152.6) Pre-operative intravenous
fluid* 46 (12.8%) 45 (12.6%) 36 (12.6%)
Hemoglobin (g.100ml)* 10.3 (2.1) 10.2 (2.0) 10.3 (2.0) Baseline oxygen saturation 99 (98-100) 99 (98-100) 99 (98-100) Baseline heart rate 152.4 (19.7) 149.9 (16.3) 153.4 (19.9) Surgical details* Bilateral hernia
exploration/repair 162 (45.9%) 161 (45.4%) 127 (44.4%)
Surgery time: from knife to
skin to last stitch (mins) 28.0 (20-38) 28 (20-40) 26.0 (19-35)
Anesthesia details* Anesthesia time: from start
skin prep for regional or
induction of GA, to out of
operating theatre (mins)
51.0 (40-69) 66 (51.5-84.5) 46.5 (39-61)
Suxamethonium 0 1 (0.3%) 0 Non depolarising 20 (5.5%) 125 (34.9%) 0
33
neuromuscular blocker Spinal without caudal* 222 (64%) 0 193 (67%) Caudal without spinal* 7 (2%) 332 (93.3%) 4 (1%) Caudal plus spinal* 117 (34%) 0 89 (31%) Ilioinguinal block 3 (0.8%) 16 (4.5%) 2 (0.7%) Field block 51 (14.4%) 40 (11.2%) 36 (12.6%) Laryngeal mask airway used 7 (2.0%) 60 (16.9%) 0 Tracheal tube used 40 (11.3%) 281 (79.4%) 0
Minimum systolic blood
pressure (mmHg)
70·7 (15·3) 54·8 (11·7) 73·2 (14·3)
Data presented as mean and standard deviation or median and interquartile range or frequencies
and percentage of non-missing data.
*Note these data refer to all cases where the listed blocks were attempted before the start of
surgery whether the blocks were effective or not. GA as per protocol data are not presented as
only two children in the GA arm had surgery cancelled, so the data is very similar to the
intention to treat data. ECG= electrocardiogram; GA= general anesthesia; IQR=interquartile
range; RA=regional anesthesia.
Table 2.
34
Intention to Treat Outcome by
frequency
and by time
period
GA (355) RA (354) As per
Protocol-
RA (286)
RA to
Supplemental
GA/Sedation (23)
RA to Full GA (37)
Number of Epochs of Hypotension <35 mm Hg Any Epoch
< 35 mm Hg
49% (175) 16%(56) 9% (27) 22% (5) 51% (19)
0 51% (180) 84% (298) 91% (259) 78% (18) 49% (18) 1-4 26% (93) 11% (39) 8% (22) 17% (4) 30% (11) 5-9 15% (54) 3% (12) 2% (5) 4% (1) 11% (4) 10-15 7% (25) 1% (3) 0% (0) 0% (0) 5% (2) 15-19 1% (3) 0% (1) 0% (0) 0% (0) 3% (1) 20 or > 0% (0) 0% (1) 0% (0) 0% (0) 3% (1)
Number of Epochs of Hypotension <45 mm Hg
Any Epoch
< 45 mm Hg 87% (309) 41%(145) 34% (98) 52% (12) 76% (28)
0 13% (46) 59% (209) 66% (188) 48% (11) 24% (9) 1-4 20% (72) 22% (79) 23% (65) 17% (4) 24% (9) 5-9 36% (128) 14% (49) 10% (29) 26% (6) 32% (12) 10-14 23% (82) 4% (13) 1% (4) 9% (2) 14% (5) 15-19 6% (20) 1% (2) 0% (0) 0% (0) 0% (0) 20 or > 2% (7) 1% (2) 0% (0) 0% (0) 5% (2)
Table 2. Outcomes: Frequency of Hypotensive Events during Anesthesia Time and Blood
Pressure Estimates by Time Period. Outcome measures are reported in percentage of total
patients from each treatment group who exhibited outcome and comparison between RA and GA
groups are presented as relative risk (RR) with 95% confidence interval (CI) and p-value. RA
number for ITT is 354 rather than 361 because of missing outcome data; there were 5 cancelled
surgeries and 1 withdrawal of consent and 1 RA case where no outcome data was collected. GA
35
number for ITT is 355 rather than 358 because there were 2 cancelled surgeries and 1 GA case
with missing hemodynamic data. a Estimate not adjusted for gestational age.
Table 3.
Estimates of MAP and HR
Time Period GA Mean
(SD)
RA
Mean (SD)
As per
Protocol-
RA Mean
(SD) (286)
Estimate
Difference GA-
RA (ITT)
95% CI
Preincision
MAP
43.7 (10.5) 55.8 (13.7) 58.2 (13.1) -12.9
[-14.5, -11.3]*
Preincision
HR
144.6 (15.2) 159.3 (21.8) 160.6
(22.1)
-15.4
[-17.6, -13.2]*
Surgical
MAP
40.9 (9.3)
52.1 (11.6) 54.2 (10.5) -12.2
[-13.6, -10.8]*
Surgical HR 138.3 (14.7) 146.8 (19.0) 147.3
(19.2)
-8.9
[-11.1, -6.7]*
Anesthesia
MAP
42.4 (10.2 53.0 (12.2) 55.2 (11.3) -11.3
[-12.7, -10.0]*
Anesthesia
HR
141.9 (16.3) 150.9 (20.8) 151.6
(21.1)
-10.3
[-12.2, -8.3]*
Table 3. Time period data presented as mean and standard deviation in mm Hg ; ITT=intention
to treat, APP= as per protocol,MAP=mean arterial pressure, GA= general anesthesia,
RA=regional anesthesia, SD=standard deviation. CI = Confidence Interval. *P value <0.001
Values are unadjusted for gestational age at birth.
36
Table 4. Relative Risks for Hypotension-related outcomes General Anesthesia compared with
Regional Anesthesia
Intention to Treat As Per Protocol
Outcome Relative
Risk (95%
CI)
P value Relative
Risk (95%
CI)
P value
Anesthesia
Time AnyHypoten
sion MAP<45 mm
Hg (single
epoch) a
2.1
[1.8,2.4]
<0.001 2.5 [2.1, 3.0] <0.001
MAP<45 mm
Hg (triple
epoch)
3.5
[2.7,4.4]
<0.001 6.8
[4.5, 10.1]
<0.001
Moderate
Hypotension MAP<35 mm
Hg (single
epoch)
2.8 [2.0,
4.1]
<0.001 4.5
[2.7, 7.4]
<0.001
MAP<35 mm
Hg (triple
epoch)
5.1
[2.7,10.1]
<0.001 44.7
[6.9,290.1 ]
<0.001
Any
Intervention for
Blood Pressureb
2.8
[1.7,4.4]
<0.001 2.8
[1.7, 4.9]
<0.001
Preincision
Time AnyHypoten
sion MAP<45 mm
Hg (single
epoch)
2.8
[2.1, 3.6]
<0.001 3.6
[2.4, 5.3]
<0.001
MAP<45 mm 12.3 <0.001 NA
37
Hg ( triple
epoch)
[4.8,31.4]
Moderate
Hypotension MAP<35 mm
Hg (single
epoch)
3.4
[2.3, 5.0]
<0.001 5.5 [3.1,9.9]
<0.001
MAP<35 mm
Hg (triple
epoch)
NA NA
Surgical
Time Any
Hypotension MAP<45 mm
Hg (single
epoch)a
2.1
[1.8, 2.5]
<0.001 2.6 [2.2,3.1]
<0.001
MAP<45 mm
Hg ( triple
epoch)
3.0
[2.3, 4.1]
<0.001 6.6
[4.4, 9.8]
<0.001
Moderate
Hypotension MAP<35 mm
Hg
3.2
[2.2, 4.7]
<0.001 5.6
[3.5, 8.9]
<0.001
MAP<35 mm
Hg (triple
epoch)
5.2
[2.4, 11.1]
<0.001 NA
Estimates of difference between groups presented as relative risk and 95% confidence intervals
(95% CI). a Estimate not adjusted for gestational age. b Estimate obtained from generalised linear
model using a log-link assuming outcome data follows a binomial distribution and robust
standard errors allowing for observations within sites to be clustered.
38
Table 5. Relative risk of 3 consecutive epochs of MAP <35 mm trivariable and multivariable
model during anesthesia time (time in the room).
Predictor Relati
ve
risk
tri*
95% CI P value Relati
ve
risk
multi
N=60
9
95%
CI
P value
General Anesthesia vs
Regional Anesthesiaa
5.62 2.60,12.1 <0.001 4.64 2.23,9.66 <0.001
Gestational birth age (days)a 0.94 .90,0.98 0.009 1.01 0.97,1.06 0.533
Weight at surgery (kg) 0.60 0.49,0.72 <0.001 0.56 0.47,0.67 <0.001
39
Log Fasting time (1-2.7, 2.7-
7.4, 7.4-20 hrs)
0.95 0.65,1.40 0.808 0.94 0.67,1.32 0.711
Log Surgery time (1-2.7, 2.7-
7.4, 7.4-20 hrs)
1.27 0.78,2.07 0.331 1.43 0.87,2.36 0.161
Sex male vs female 1.05 0.59,1.85 0.873 0.96 0.56,1.63 0.870
Chronologic Age at surgery
(weeks)b
0.92 0.86,0.97 0.003 - - -
PMA at surgery (weeks)b 0.92 0.86,0.97 0.003 - - -
Mean Temperature (degrees
Celsius)c
0.73 0.51,1.03 0.073 - - -
Minimum Temperature
(degrees Celsius)
0.74 0.60,0.90 0.003 0.78 0.65,0.95 0.011
Estimates of difference between groups presented as relative risk and 95% confidence intervals
(95% CI). Relative risk tri is the ratio of risk for factor of interest plus randomized anesthetic
group and gestational age, and includes all non-missing observations. Relative risk multi is the
ratio of risk found in a multivariable model. a Bivariable model containing group and gestational
age as predictors b Factors were strongly associated with weight at surgery and therefore not
included in full model. c Factor was strongly associated with mean temperature and therefore not
included in full model. PMA means post menstrual age.
40
Table 6. Interventions for Hypotension
41
Intervention GA RA (ITT) RA (APP)
Bolus only 14.4% (51) 5.6% (20) 5.2% (15)
Vasoactive only 2.8% (10) 1.4% (5) 1.4% (4)
Both 2.0% (7) 0.3% (1) 0 (0)
Total 19.2% (68) 7.3% (26)
6.6% (19)
Data presented as percentage of group that was treated. Parentheses represent actual number of
patients treated. ITT=intention to treat, APP= as per protocol. GA=general anesthesia,
RA=regional anesthesia.